Sustainable Diets and Meat Consumption Reduction in Emerging Economies: Evidence from Colombia

: The growing demand for meat and animal products in emerging economies has become a concern given its environmental and health impacts. The sustainable diets approach has emerged to address the multidimensional challenge of reaching a context-based diet that minimizes negative environmental impacts, provides health and nutrition to all segments of the population, and is a ﬀ ordable and coherent with the local culture and traditions. The aim of this study was to explore the prospects for meat consumption reduction and challenges encompassing the environmental, and health spheres. In order to do so, we analyzed: (1) The current carbon and water per capita footprints for two animal-based options and two plant-based options; and (2) the contribution of each food alternative to the local dietary reference intakes based on average per capita daily consumption and signiﬁcant di ﬀ erences among the nutrient values for each food alternative through a two proportion Z-test. Our results show that the annual per capita carbon and water footprints for beef were higher compared to other alternatives, despite a higher per capita consumption of chicken. Also, our ﬁndings reveal that the average consumption of beef and chicken contributes 39% of the maximum recommended daily intake for cholesterol and 61% of the Recommended Dietary Allowance for protein in the country. Finally, relevant promoting forces and barriers related to meat consumption reduction were identiﬁed based on the two dimensions evaluated. This study calls for a joint e ﬀ ort to make changes in public policy, food systems, and consumer education.


Introduction
Despite increasing information about human-induced climate change and the nutritional challenges that impact human health, most consumers are not aware of the environmental and health impacts of their everyday food choices [1][2][3], and even if they are, there are several barriers that might route their consumption decisions toward less sustainable options [1,4]. It is estimated that by 2050, the world's population will reach 9.6 billion, and some argue that farmers around the world will have to double the current agricultural production in order to produce enough food to keep up with the trend toward a meat-centric diet [5]. Meat consumption has tripled in the past 50 years, and although it is no longer increasing in well-established economies, meat demand is still growing in emerging economies [6].
This trend has become a concern due to its negative environmental impacts [7,8] and is also considered controversial from a health perspective [9]. However, at the socio-economic level, meat consumption is generally perceived in a positive way and has become an encouraged habit in some emerging economies in order to guarantee the appropriate intake of some macro and micronutrients in the population [10,11]. The sustainable diets approach has emerged to address the multidimensional In order to undertake an integrative analysis, we used the nutrition ecology framework [31]. This framework allows the assessment of complex nutrition problems from a multidimensional perspective, increasing the level of integration in research and providing the basis for scientific policy advice. Through this framework, it is possible to analyze the interrelatedness of nutritional issues by encompassing dimensions, such as health, environment, society, and economy, simultaneously. Thus, by analyzing multiple dimensions, it is possible to get a wider perspective of the problem and acknowledge the emerging tensions [31]. Given the scope of this study, we focused our analysis on the health/nutrition and environmental dimensions, opening the possibility of further research on the social and economy dimensions.
In this study, we assessed the environmental and health/nutrition dimensions of the consumption of two animal-based protein sources and two plant-based protein sources. Beef and poultry were chosen since these two types of meat account for almost 75% of the meat consumption in the country [28]. Since red beans and lentils are the most consumed legumes in Colombia [32], these foods were also included to provide a contrast with animal-based options.
This study was based on desk research. The desk research consisted of calculating the nutritional and environmental impacts of the four options. In order to do this, several estimations were made. Given the fact that this is an under-researched area, we used official documents when available, research results from similar contexts such as Brazil, and reports from producers' organizations. Below, we explain in detail the procedures by which we calculated each dimension.

Methods for the Environmental Dimension
First, environmental impacts were calculated based on the following data: (1) The annual average consumption per capita of beef, chicken, red beans, and lentils in Colombia [28]; (2) the water footprint calculated by Hoekstra and Mekonnen [33] for specific production systems and locations; and (3) the carbon footprint related to the production systems and locations [34][35][36][37].
Calculations of the carbon footprint of beef, poultry, red beans, and lentils included carbon emissions per kg of product in the production phase and post-farm gate phase, as well as the average per capita consumption of each product in Colombia. Considering the lack of information about local production systems and post-farm gate conditions, the calculations for this study were based on data available about similar conditions in other countries, and estimations of individual footprints included data of the average local consumption. Thus, beef production emissions were based on information available from grazing systems in Brazil [36], chicken production emissions were based on industrial systems in Canada (given the similarities in the slaughter weight and diet), and the red bean and lentil production emissions and the post-farm gate carbon emissions for the four products were based on data from the U.S. and Canada [34,35].
Since beef imports in Colombia only represent 0.06% of the total national production [38], and chicken imports only represent 2.2% [39], it was assumed in the calculations of the environmental impacts that all the beef and poultry consumed was locally produced. It was also assumed that all the beef consumed was produced under a grazing production system, and that poultry was produced under an industrial system. Additionally, given that almost 80% of the red beans consumed in 2018 in the country were locally produced, and 80% of lentils were imported from Canada [40], this project's calculations assumed that all the red beans were produced in Colombia, and all the lentils were imported from Canada (via international water shipping containers). The limitations related to these estimations and assumptions are discussed in Section 5.

Methods for the Health/Nutrition Dimension
Second, the nutritional value per 100 g of the food products mentioned above (beef, poultry, red beans, and lentils) provided by the ICBF Colombian Food Composition Tables [41] (ICBF is the acronym for the Colombian Family Welfare Institute), the Energy and Nutrient Recommendations of the Ministry of Health [42], and the average per capita consumption of these four products in the Sustainability 2019, 11, 6595 4 of 17 country were analyzed with the purpose of identifying key nutritional factors per alternative. Since loin is the most popular beef cut among consumers in Colombia [43,44], chicken breasts are the most popular poultry cut [45], and red beans and lentils are the most consumed legumes in the country [32], the nutritional analysis presented in this project corresponds to the information provided by the ICBF Colombian Food Composition Tables [41] for: Chicken breast, skinless, cooked, no salt added; beef, loin, cooked, no salt added; red beans, cooked no salt added; and lentils, cooked, no salt added.

Statistical Analysis
We analyzed significant differences among the proportion of nutrients per 100 g of each product through a two proportion Z-test (R Studio Version 1.0.136). Based on the information available in the Colombian Food Composition Tables [41], the proportions of the macro and micronutrients of beef, chicken, red beans, and lentils were compared. The null hypothesis in each test assumed that the proportions were the same. The alternative hypothesis was that the proportions were different (two-sided test). We used a confidence level of 95% and a confidence interval clipped to [−1,1]. We obtained the Z statistic by taking the squared root of the Chi-squared statistic that comes by default in the function. The p-values obtained are specified in Table 1.

Environmental Dimension
In 2018, in Colombia, per capita annual consumption of the protein types studied were: 18.2 kg of beef, 33.8 kg of chicken [28], 3.2 kg of red beans, and 1.5 kg of lentils [40].

Water Footprint
According to the data provided by Hoekstra and Mekonnen [33], the water footprint of beef produced in Colombia under a grazing system corresponds to 7545 m 3 /ton [33]; poultry produced under an industrial system in Colombia corresponds to 3246 m 3 /ton of water [33]; red beans produced in Colombia equals 2671 m 3 /ton of water [46]; and the global average water footprint for lentils is 5874 m 3 /ton [46]. Thus, taking into account the average per capita consumption of these products in Colombia, as well as the estimated water footprint presented above, beef had the highest water footprint among the four food types (137,319 lt per person per year), although per capita chicken consumption was higher in the country. The annual water footprint per capita per product is described in Figure 1.

Carbon Footprint
The calculations of the carbon footprint of beef, poultry, red beans, and lentils included carbon emissions per kg of product in the production phase and post-farm gate phase. Production emissions comprised all the "emissions before the product leaves the farm" [35] (p. 23). Post-farm gate emissions encompassed the estimated data per product of processing, domestic transport, refrigeration, home cooking, and disposal [35]. Since there is not enough data available about specific carbon emissions related to these processes in Colombia, the calculations for this study were based on data available about the production emissions in Brazil [36], and the post-farm gate carbon emissions from the U.S. and Canada [34,35].
Given the similarities between Colombia and Brazil regarding beef production, the carbon footprint used was 28 kg of CO2eq emissions per kg of carcass weight at the production phase (in a scenario that does not include land use change), and 44 kg of CO2eq per kg of carcass weight (in a scenario that includes Land Use Change) [36]. Post-farm gate CO2eq emissions for beef corresponded to 3.7 kg of CO2 per kg of consumed beef [35]. Thus, the total carbon footprint of beef, not including LUC, was 31.7 kg of CO2eq per kg of carcass weight, and including LUC was 47.7 kg CO2eq per kg of carcass weight. Beef had the highest carbon footprint of the food types compared in this study.
The Environmental Working Group [35] estimated that poultry production under an industrial system emits 3.6 kg CO2eq per kg of edible chicken. This production system was assumed to be the same in Colombia and Canada, considering the similarities in the slaughter weight (2300-2750 g), and a similar diet of corn and soy-based concentrate [35,47]. Post-farm gate emissions corresponded to 3.3 kg CO2eq per kg of consumed chicken. The total carbon footprint of poultry in Colombia, including production and post-farm gate emissions, was estimated to be 6.9 kg of CO2eq per kg of product [35].
In contrast, vegetable protein alternatives, such as red beans and lentils, had a lower carbon footprint. According to the Environmental Working Group [34], the CO2 emissions of these food types are mainly linked to the energy required to cook them. It was estimated that the emissions from the production of red beans corresponded to 1.3 kg of CO2eq per kg of product, and the post-farm gate emissions corresponded to 0.7 kg of CO2eq per kg of product, for a total carbon footprint of 2 kg of CO2eq per kg of product [34]. In the case of lentils, the production phase represented 0.37 kg of CO2eq per kg of product. Since lentils are imported from Canada, the shipping-related emissions were calculated based on the estimates of Weber and Matthews [37], and added to the post-farm gate value, which then led to a calculated result for this phase of 0.63 kg of CO2eq per kg of product. Thus, the total calculated carbon footprint for lentils was 1 kg of CO2eq per kg of product. The estimations of water and carbon footprints are presented in Table 2.

Carbon Footprint
The calculations of the carbon footprint of beef, poultry, red beans, and lentils included carbon emissions per kg of product in the production phase and post-farm gate phase. Production emissions comprised all the "emissions before the product leaves the farm" [35] (p. 23). Post-farm gate emissions encompassed the estimated data per product of processing, domestic transport, refrigeration, home cooking, and disposal [35]. Since there is not enough data available about specific carbon emissions related to these processes in Colombia, the calculations for this study were based on data available about the production emissions in Brazil [36], and the post-farm gate carbon emissions from the U.S. and Canada [34,35].
Given the similarities between Colombia and Brazil regarding beef production, the carbon footprint used was 28 kg of CO 2 eq emissions per kg of carcass weight at the production phase (in a scenario that does not include land use change), and 44 kg of CO 2 eq per kg of carcass weight (in a scenario that includes Land Use Change) [36]. Post-farm gate CO 2 eq emissions for beef corresponded to 3.7 kg of CO 2 per kg of consumed beef [35]. Thus, the total carbon footprint of beef, not including LUC, was 31.7 kg of CO 2 eq per kg of carcass weight, and including LUC was 47.7 kg CO 2 eq per kg of carcass weight. Beef had the highest carbon footprint of the food types compared in this study.
The Environmental Working Group [35] estimated that poultry production under an industrial system emits 3.6 kg CO 2 eq per kg of edible chicken. This production system was assumed to be the same in Colombia and Canada, considering the similarities in the slaughter weight (2300-2750 g), and a similar diet of corn and soy-based concentrate [35,47]. Post-farm gate emissions corresponded to 3.3 kg CO 2 eq per kg of consumed chicken. The total carbon footprint of poultry in Colombia, including production and post-farm gate emissions, was estimated to be 6.9 kg of CO 2 eq per kg of product [35].
In contrast, vegetable protein alternatives, such as red beans and lentils, had a lower carbon footprint. According to the Environmental Working Group [34], the CO 2 emissions of these food types are mainly linked to the energy required to cook them. It was estimated that the emissions from the production of red beans corresponded to 1.3 kg of CO 2 eq per kg of product, and the post-farm gate emissions corresponded to 0.7 kg of CO 2 eq per kg of product, for a total carbon footprint of 2 kg of CO 2 eq per kg of product [34]. In the case of lentils, the production phase represented 0.37 kg of CO 2 eq per kg of product. Since lentils are imported from Canada, the shipping-related emissions were calculated based on the estimates of Weber and Matthews [37], and added to the post-farm gate value, which then led to a calculated result for this phase of 0.63 kg of CO 2 eq per kg of product. Thus, the total calculated carbon footprint for lentils was 1 kg of CO 2 eq per kg of product. The estimations of water and carbon footprints are presented in Table 2. Based on the data presented above and taking into account the average per capita consumption in 2018 in Colombia, beef had the highest carbon footprint with 577.3 kg of CO 2 eq per person per year not including LUC, and 868.5 kg of CO 2 eq per person per year when including LUC, which is summarized in Figure 2.  Based on the data presented above and taking into account the average per capita consumption in 2018 in Colombia, beef had the highest carbon footprint with 577.3 kg of CO2eq per person per year not including LUC, and 868.5 kg of CO2eq per person per year when including LUC, which is summarized in Figure 2.

Health/Nutrition Dimension
Calculations for the health dimension allowed a comparison among the nutritional contributions of the four protein alternatives analyzed, considering the average annual per capita consumption [28,40]. Table 3 summarizes the nutritional values of each alternative based on the Colombian Food Composition Table [41] and the Energy and Nutrient Recommendations of the Ministry of Health for Colombian men and women with an average weight of 60 to 70 kg, between 18 and 59 years old, and with a low/moderate level of physical activity [42]. Table 1 presents the statistical analysis of the nutrient values for each product. Our findings show that, due to a low per capita consumption of the plant-based options studied, the nutritional contributions of red beans and lentils to the average diet in Colombia are currently small, representing approximately 1.4% of the recommended daily allowance (RDA) of protein that is suggested by Colombia's Ministry of Health as part of a healthy diet [42]. Animal food alternatives, such as beef and chicken, have a significantly higher (p < 0.05) content of protein compared to red beans and lentils, and are consumed in higher amounts by the population, representing 61% of the RDA.
Regarding total fat, beef has a significantly higher value (p < 0.05) compared with the two plant-based options analyzed, and no significant differences (p < 0.05) with chicken. The total fat value per 100 g of chicken had no significant differences (p < 0.05) with red beans or lentils. The average per capita consumption of chicken and beef provides 19% of the maximum recommended daily intake of total fat while the average per capita consumption of red beans and lentils represents less than 0.3%. The average per capita consumption of the animal-based proteins analyzed also provides 24% of the maximum recommended daily intake of saturated fats, and 75% of the recommended intake of polyunsaturated fatty acids. Additionally, we found that the value for cholesterol is significantly higher (p < 0.05) in the animal-based foods analyzed. According to our calculations, the average per capita consumption of chicken and beef provides approximately 39% of the maximum recommended daily intake of cholesterol. The value of cholesterol, saturated, monounsaturated, and polyunsaturated fatty acids for red beans and lentils was not available in local sources, however, other sources [48,49] report that 100 g of red beans and lentils have less than 1% of the maximum recommended daily intake of saturated fats, and no cholesterol content.
The value of carbohydrates and dietary fiber per 100 g of the two plant-based alternatives was significantly higher (p < 0.05) than the value of these nutrients per 100 g of beef and chicken. The average per capita consumption of red beans and lentils accounts for 2.4% of the RDA for carbohydrates and 3% of the adequate intake (AI) of dietary fiber. Meanwhile, since the content of carbohydrates and fiber of the two animal-based alternatives evaluated is minimal, the average consumption of these two products does not contribute significantly to this macronutrient.
Regarding minerals, when comparing 100 g of each product, our results show a higher content of magnesium, calcium, and potassium in red beans versus the other three alternatives (p < 0.05). Also, we found a higher content of phosphorus in beef compared with the plant-based options analyzed and a lower content of phosphorus when comparing lentils with the other three alternatives (p < 0.05). We found a significantly higher content of sodium in 100 g of the animal-based alternatives versus the content of sodium in 100 g of the plant-based options (p < 0.05). No statistically significant differences were found in the values per 100 g of iron and zinc among the four products; however, the value of zinc per 100 g of beef is closer to the RDA [42]. The average per capita consumption of beef and chicken in Colombia provides 12% of the RDA for iron, 10% of the RDA for magnesium, 36% of the RDA for phosphorus, 8% of the RDA for potassium, 5% of the RDA for sodium, and 37% of the RDA for zinc. Meanwhile, given a lower per capita consumption of red beans and lentils, the contributions of these foods to the RDA for each of the above-mentioned minerals is less than 3%. In the case of iron, the average availability in the country has reached 93%, thanks to the increased availability of meat and animal products [33]. Zinc deficiency is one of the most common micronutrient deficiencies in children in Colombia, with a prevalence in 43% of the population between 1 and 4 years of age [11]. Data regarding nutritional contributions based on food availability [29] suggest that local per capita consumption of phosphorus exceeds daily recommendations in the country, making it relevant to reduce phosphorus consumption due to its association with increased all-cause mortality [50].
As for vitamins, our results show a significantly higher content of niacin per 100 g of chicken when compared to 100 g of the plant-based options evaluated, but no significant differences with the content of niacin in 100 g of beef loin (p < 0.05). We found a significantly higher content of folate in 100 g of the plant-based alternatives versus the content of folate in 100 g of the animal-based options, with red beans having a higher content of this nutrient when comparing it to the other three alternatives (p < 0.05). Regarding vitamin A, our results show that 100 g of chicken have a higher content of this nutrient when compared with beef or red beans but not with lentils (p < 0.05). No statistically significant differences were found in the values per 100 g of vitamin C, thiamin, riboflavin, and vitamin B12 (p < 0.05). Notwithstanding, it is relevant to consider that 100 g of beef loin provides a quantity of vitamin B12 that exceeds the RDA. The average per capita consumption of the two animal-based alternatives provides 7% of the RDA for thiamin, 19% of the RDA for riboflavin, 67% of the RDA for niacin, 2% of the RDA for folate, 66% of the RDA for vitamin B12, and less than 1% of the RDA for vitamin A. The average per capita consumption of the two plant-based alternatives evaluated provides 3% of the RDA for thiamin, 1% of the RDA for riboflavin, 1% of the RDA for niacin, 4% of the RDA for folate, and has no contributions to the RDA for vitamin B12 and vitamin A. In Colombia, 18.1% of the population between 5 and 12 years old is at risk of vitamin B12 deficiency, and 13.2% of women between 13 and 49 years old present with vitamin B12 deficiency [11]. Figure 3 shows the contributions of each food alternative to the recommended daily value for nutrients. with red beans having a higher content of this nutrient when comparing it to the other three alternatives (p < 0.05). Regarding vitamin A, our results show that 100 g of chicken have a higher content of this nutrient when compared with beef or red beans but not with lentils (p < 0.05). No statistically significant differences were found in the values per 100 g of vitamin C, thiamin, riboflavin, and vitamin B12 (p < 0.05). Notwithstanding, it is relevant to consider that 100 g of beef loin provides a quantity of vitamin B12 that exceeds the RDA. The average per capita consumption of the two animal-based alternatives provides 7% of the RDA for thiamin, 19% of the RDA for riboflavin, 67% of the RDA for niacin, 2% of the RDA for folate, 66% of the RDA for vitamin B12, and less than 1% of the RDA for vitamin A. The average per capita consumption of the two plant-based alternatives evaluated provides 3% of the RDA for thiamin, 1% of the RDA for riboflavin, 1% of the RDA for niacin, 4% of the RDA for folate, and has no contributions to the RDA for vitamin B12 and vitamin A. In Colombia, 18.1% of the population between 5 and 12 years old is at risk of vitamin B12 deficiency, and 13.2% of women between 13 and 49 years old present with vitamin B12 deficiency [11]. Figure 3 shows the contributions of each food alternative to the recommended daily value for nutrients.

Discussion
This study adds to the discussion on sustainable diets by exploring the environmental and health/nutrition dimensions of meat consumption reduction in an emerging economy. Through the nutrition ecology framework, it was possible to address this complex issue in an integrative manner [31]. To the best of the author's knowledge, this is the first time this has been explored in an emerging economy, characterized by a growing middle class that consumes increasing amounts of animal protein, at the same time that key ecosystems are put at risk. An increased consumption of animal food products and the transition towards a diet high in animal fats and low in fruits and vegetables in the country calls for the promotion of sustainable food choices and diets. This, according to the literature, should focus on the reduction in meat consumption [7,10,[51][52][53]. The nutrition ecology framework [31] allowed an integrative approach to the possible impacts of a reduction in meat consumption in the country, and permitted joining the discussion about sustainable diets and consumption choices [54] (see Figure 4). In this section, we analyze the promoting forces and barriers of beef consumption reduction in the country.

Discussion
This study adds to the discussion on sustainable diets by exploring the environmental and health/nutrition dimensions of meat consumption reduction in an emerging economy. Through the nutrition ecology framework, it was possible to address this complex issue in an integrative manner [31]. To the best of the author's knowledge, this is the first time this has been explored in an emerging economy, characterized by a growing middle class that consumes increasing amounts of animal protein, at the same time that key ecosystems are put at risk. An increased consumption of animal food products and the transition towards a diet high in animal fats and low in fruits and vegetables in the country calls for the promotion of sustainable food choices and diets. This, according to the literature, should focus on the reduction in meat consumption [7,10,[51][52][53]. The nutrition ecology framework [31] allowed an integrative approach to the possible impacts of a reduction in meat consumption in the country, and permitted joining the discussion about sustainable diets and consumption choices [54] (see Figure 4). In this section, we analyze the promoting forces and barriers of beef consumption reduction in the country. Sustainability 2019, 11, x FOR PEER REVIEW 3 of 18 On the one hand, our results show that reducing per capita beef consumption in 33%approximately one portion less of beef per week-and replacing it with any of the plant-based alternatives studied (red beans or lentils) could reduce environmental impacts and represent health benefits. Given that beef had the highest carbon and water footprints [55,56], replacing one portion per week of meat with a portion of chicken, red beans, or lentils could represent a reduction in the annual per capita water footprint of Colombians between 9682 and 29,113 lt, as well as a reduction in the annual per capita carbon footprint between 245 and 281 kg of CO2eq (with LUC) (see Table 4). These figures are equivalent to the amount of water used by an average family in Colombia between 23 and 68 days [57] or taking a car off the road for 976 to 1118 km [58]. Additionally, our results show that replacing one portion of beef per week (approximately 6.1 kg per person per year) with the plant-based alternatives studied (red beans and lentils) could represent a lower intake of total fat and cholesterol (see Figure 5). Infarction-or cardiac ischemiaand cerebrovascular disease are two leading causes of death in Colombia, and both diseases are linked to unhealthy lifestyles and high levels of blood cholesterol [59]. Research shows that a reduction in the consumption of animal meat and other animal products, along with an increase in the consumption of plant-based foods, have important health benefits, such as lowering the rate of  On the one hand, our results show that reducing per capita beef consumption in 33%approximately one portion less of beef per week-and replacing it with any of the plant-based alternatives studied (red beans or lentils) could reduce environmental impacts and represent health benefits. Given that beef had the highest carbon and water footprints [55,56], replacing one portion per week of meat with a portion of chicken, red beans, or lentils could represent a reduction in the annual per capita water footprint of Colombians between 9682 and 29,113 lt, as well as a reduction in the annual per capita carbon footprint between 245 and 281 kg of CO 2 eq (with LUC) (see Table 4). These figures are equivalent to the amount of water used by an average family in Colombia between 23 and 68 days [57] or taking a car off the road for 976 to 1118 km [58]. Additionally, our results show that replacing one portion of beef per week (approximately 6.1 kg per person per year) with the plant-based alternatives studied (red beans and lentils) could represent a lower intake of total fat and cholesterol (see Figure 5). Infarction-or cardiac ischemia-and cerebrovascular disease are two leading causes of death in Colombia, and both diseases are linked to unhealthy lifestyles and high levels of blood cholesterol [59]. Research shows that a reduction in the consumption of animal meat and other animal products, along with an increase in the consumption of plant-based foods, have important health benefits, such as lowering the rate of cardiac events and even reversing Cardio Vascular Disease [60], as well as reducing the risk of having ischemic heart disease [61]. Also, our results show that a higher consumption of plant-based alternatives, such as red beans, could increase the intake of nutrients, such as dietary fiber, potassium, and folate. cardiac events and even reversing Cardio Vascular Disease [60], as well as reducing the risk of having ischemic heart disease [61]. Also, our results show that a higher consumption of plant-based alternatives, such as red beans, could increase the intake of nutrients, such as dietary fiber, potassium, and folate. On the other hand, some barriers to meat consumption reduction were also identified. Even though there is clear evidence of the environmental footprint and health impacts of animal-based diets [7,62] in Colombia, reducing meat consumption could face other obstacles that should be taken into account, such as the availability of the products that could substitute meat. While Colombia is self-sufficient in terms of fruits, roots, tubers, sugar, meat, and animal products, the country depends on the imports of legumes, such as lentils and chickpeas [29].
It is also important to consider that lean meat cuts, such as the ones evaluated in this study, provide important nutrients, such as protein, zinc, and vitamin B12, to the Colombian population given their availability [29]. Beef consumption reduction could represent a risk of deficiencies in these nutrients if a variety of plant-based foods are not taken, other animal products are not consumed, or other foods are not fortified, as suggested by Esselstyn et al. [60]. Particularly, in the case of protein, the scenario proposed in this study-a 33% beef consumption reduction and replacement with plantbased protein-could lower the RDA for protein by 11% if no other changes are made to the diet (see Figure 5).
However, it is worth noting that the local Nutrient Recommendations of the Ministry of Health [42] state that the RDA for protein should be 1.1 g of protein per kg of weight per day, which is surprisingly higher than the RDA for protein in the United States [63], Canada [64], and Brazil [65]. In these countries, this value corresponds to 0.8 g of protein per kg of weight per day. The nutritional value per 100 g of beef loin provided by the ICBF Colombian Food Composition Tables [44] corresponds to 36.1, which is also higher than the values in the United States (28.57 g) [48], Canada (28.79 g) [66], and Brazil (31.9 g) [67]. Moreover, the National Nutritional Guidelines [68] explicitly recommend that the population between 18 and 59 years of age should consume 99.5 g of protein per day for men, and 78.7 g for women, and suggest that approximately one third of a healthy meal should consist of animal products [68]. Thus, in order to meet a higher RDA, meat consumption is validated and promoted through official documents. Nonetheless, in order to improve citizens' and the planet's health, it has been argued that it is relevant to promote sustainable diets through national guidelines, which would require changes in policy frameworks regarding food production and On the other hand, some barriers to meat consumption reduction were also identified. Even though there is clear evidence of the environmental footprint and health impacts of animal-based diets [7,62] in Colombia, reducing meat consumption could face other obstacles that should be taken into account, such as the availability of the products that could substitute meat. While Colombia is self-sufficient in terms of fruits, roots, tubers, sugar, meat, and animal products, the country depends on the imports of legumes, such as lentils and chickpeas [29].
It is also important to consider that lean meat cuts, such as the ones evaluated in this study, provide important nutrients, such as protein, zinc, and vitamin B12, to the Colombian population given their availability [29]. Beef consumption reduction could represent a risk of deficiencies in these nutrients if a variety of plant-based foods are not taken, other animal products are not consumed, or other foods are not fortified, as suggested by Esselstyn et al. [60]. Particularly, in the case of protein, the scenario proposed in this study-a 33% beef consumption reduction and replacement with plant-based protein-could lower the RDA for protein by 11% if no other changes are made to the diet (see Figure 5).
However, it is worth noting that the local Nutrient Recommendations of the Ministry of Health [42] state that the RDA for protein should be 1.1 g of protein per kg of weight per day, which is surprisingly higher than the RDA for protein in the United States [63], Canada [64], and Brazil [65]. In these countries, this value corresponds to 0.8 g of protein per kg of weight per day. The nutritional value per 100 g of beef loin provided by the ICBF Colombian Food Composition Tables [44] corresponds to 36.1, which is also higher than the values in the United States (28.57 g) [48], Canada (28.79 g) [66], and Brazil (31.9 g) [67]. Moreover, the National Nutritional Guidelines [68] explicitly recommend that the population between 18 and 59 years of age should consume 99.5 g of protein per day for men, and 78.7 g for women, and suggest that approximately one third of a healthy meal should consist of animal products [68]. Thus, in order to meet a higher RDA, meat consumption is validated and promoted through official documents. Nonetheless, in order to improve citizens' and the planet's health, it has been argued that it is relevant to promote sustainable diets through national guidelines, which would require changes in policy frameworks regarding food production and consumption [13]. This study suggests that Colombia's current nutritional guidelines [68] are unsustainable given that they ignore the environmental dimension of food consumption, making it important to include this key component of a sustainable diet [12].
How to achieve a reduction in meat consumption and replacement with plant-based alternatives in the country is yet to be discussed. The findings of de Boer et al. [52] uncover a positive attitude towards the "meatless days" in the Netherlands, especially when consumers have clear plant-based alternatives. This could be the case for some Colombian consumers, particularly if plant-based alternatives are available and coherent with the traditions and habits of the population. Nevertheless, for this to be achieved, a significant effort is required at the level of public policy, institutions, food systems, and consumer education.
In order to achieve the positive outcomes of meat consumption reduction, an enormous institutional effort must be made, focused on: Enabling a more varied, locally available, and nutritionally rich plant-based diet and guaranteeing access to it; consumer education regarding the nutritional value of plant-based foods; and a reassessment of the national dietary guidelines in order to acknowledge the three sustainability dimensions.
The present study had clear limitations; however, the findings show clear patterns and challenges that hinder Colombians from adopting sustainable diets. In the analysis performed for the environmental dimension, Colombia's available information was limited; hence, it was necessary to use data available for Brazil, the U.S, and Canada. The data used was based on the averages provided by different sources; thus, the estimations included in this study were subject to uncertainties, which are difficult or impossible to quantify. Emissions from food loss and wastage were not included, and other environmental impacts, such as water pollution and biodiversity loss, were not calculated due to the lack of local information. In the health and nutrition dimension, only the most popular cuts and preparations of beef and poultry were considered. The exact relationship between meat consumption and chronic diseases in the country is not yet established. Additionally, social, economic, and animal welfare issues were not included in this project but could be explored in further research.
The findings of this study, as well as the framework used [31], open the path towards the exploration of socio-individual perceptions related to meat consumption reduction and replacement with plant-based alternatives. A further analysis of the social, health, economic, and environmental perceptions of consumers regarding the animal-based and plant-based options evaluated in the present study could provide a more integral approach to this issue.

Conclusions
This study explored the possibility of shifting towards a more sustainable diet by reducing meat consumption in an emerging economy, and sought to scientifically inform policy about the emerging challenges of meat consumption reduction, when considering the environmental and health spheres of this complex issue. By using a multidimensional approach, it was possible to uncover the local carbon and water impact, health, and nutritional implications of meat consumption, as well as the promoting forces and barriers for meat consumption reduction. Our findings show that reducing per capita beef consumption could reduce environmental impacts and represent health benefits. However, we identified some relevant barriers preventing this change, regarding the availability of plant-based protein sources in the country, the emergence of possible protein and nutrient deficiencies, and the recommendations by official documents and government entities about high animal-based protein consumption.